EasyDeepLearn

UMAP vs t-SNE — practical differences.

medium

Answer

  • UMAP: faster (O(n114)  vs  O(n  log  n)  with  BarnesHut  but  higher  constant)(O(n^{1}14)\;\mathrm{vs}\;O(n\;\operatorname{log}\;n)\;\mathrm{with}\;\mathrm{Barnes} - \mathrm{Hut}\;\mathrm{but}\;\mathrm{higher}\;\mathrm{constant}), preserves more global structure (inter-cluster distances less-meaningful but somewhat interpretable), supports supervised / semi-supervised variants, can transform new points (t-SNE cannot).
  • Both are for visualization primarily.
  • UMAP has become the default for 2D projection in scRNA-seq, embeddings inspection, exploratory analysis.
Check yourself — multiple choice
  • Same
  • UMAP faster + preserves more global structure + supports supervised variants + transform new points; t-SNE only fits, seed-sensitive
  • Random
  • t-SNE always better

UMAP: faster, more global structure, transforms new points, supervised variants.

#dimensionality-reduction#visualization

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions